Original title: Influence of Metric on Classification Error of Distance-Based Classifiers
Authors: Jiřina, Marcel
Document type: Research reports
Year: 2014
Language: eng
Series: Technical Report, volume: V-1211
Abstract: Five types of classifiers that use sample distances for class estimation of an unknown sample was tested. Each classifier was tested with fifteen different metrics on 24 classification tasks from the UCI Machine Learning Repository. The metrics were compared and the best of them was found for each classifier. Surprisingly, the best metrics for all five types of classifiers is the Hassanat metrics. Classifiers were also compared and ranked according to their classification ability. Wilcoxon Test and Friedman Aligned test were used for statistical evaluation.
Keywords: classifier; distance; Hassanat metrics; IINC; k-NN; metrics; multidimensional data
Rights: This work is protected under the Copyright Act No. 121/2000 Coll.

Institution: Institute of Computer Science AS ČR (web)
Original record: http://hdl.handle.net/11104/0240014

Permalink: http://www.nusl.cz/ntk/nusl-178137


The record appears in these collections:
Research > Institutes ASCR > Institute of Computer Science
Reports > Research reports